arXiv:astro-ph/0608230v1 10 Aug 2006 otatt h ainlSineFoundation. Science National the to contract TOI09mPorm–Tet e ebr fteRECONS the of Members New Twenty – Program m 0.9 CTIOPI 1 iiigAtooe,CroTll ne-mrcnOsraoy C Observatory. Inter-American Tololo Cerro Astronomer, Visiting odJ Henry J. Todd r O05+62(3.84 0253+1652 pre rough SO system perhaps are nearest than 10 three (other The within parallaxes sample. trigonometric systems RECONS previous 20 the of for horizon parallaxes the , trigonometric definitive first the B(3.85 AB otn htmti n pcrsoi bevtoshv enm stellar been 100 have nearest observations the spectroscopic among and rank photometric here porting reported systems the of h oa egbrodXI:Prla eut rmthe from Results Parallax XVII: Neighborhood Solar The hnycaagueu [email protected],subasavage@c [email protected], [email protected], srmti esrmnsfr2 e wr ytm r presen are systems dwarf red 25 for measurements Astrometric ± .2p,2t) n H 73(5.32 1723 LHS and 24th), pc, 0.02 1 eri tt nvriy tat,G 30302-4106 GA Atlanta, University, State Georgia [email protected],[email protected] [email protected], e-hnJao Wei-Chun , nvriyo igna hrotsil,V 22903 VA Charlottesville, Virginia, of University nvria eCie atao Chile Santiago, Chile, de Universidad dad Costa Edgardo ± [email protected] 0Pre Sample 10 .4p,te2r ers ytm,SR1845-6357 SCR system), nearest 23rd the pc, 0.04 [email protected] 1 onP Subasavage P. John , hlpA Ianna A. Philip ABSTRACT cetdt the to accepted 1 and e´ .M´endez Ren´e A. , 1 ± .4p,5t) nttl seven total, In 56th). pc, 0.04 srnmclJournal Astronomical 1 n hmsD Beaulieu D. Thomas and , I soeae yAR,Ic under Inc. AURA, by operated is TIO 1 iiayefforts) liminary hara.gsu.edu, d oprovide to ade ytm.Sup- systems. e,including ted, hthdno had that s co the of pc 1Jl 2006 July 31 1 –2–

full characterization of the systems, including complete VRIJHKs photometry and spectral types. A study of the variability of 27 targets reveals six obvious variable , including GJ 1207, for which we observed a flare event in the V band that caused it to brighten by 1.7 mag. Improved parallaxes for GJ 54 AB and GJ 1061, both important members of the 10 pc sample, are also reported. Definitive parallaxes for GJ 1001 A, GJ 633, and GJ 2130 ABC, all of which have been reported to be within 10 pc, indicate that they are beyond 10 pc. From the analysis of systems with (previously) high trigonometric parallax errors, we conclude that parallaxes with errors in excess of 10 mas are insufficiently reliable for inclusion in the RECONS sample. The cumulative total of new additions to the 10 pc sample since 2000 is now 34 systems — 28 by the RECONS team and six by other groups. This total represents a net increase of 16% in the number of stellar systems reliably known to be nearer than 10 pc.

Subject headings: surveys — — stars: distances — stars: low mass, brown dwarfs — stars: statistics — solar neighborhood

1. Introduction

Trigonometric parallax determinations provide one of the most fundamental measures of the cosmos. Their beauty lies in their simplicity: they are based only on geometric techniques and are the straightforward result of the Earth orbiting the Sun. Trigonometric parallaxes help define solar neighborhood membership, allow for accurate masses calculations in binary systems, provide distances to fundamental stellar clusters, and supply benchmarks for stellar population studies in the Milky Way and nearby . With concerted effort, trigonometric parallaxes with accuracies of ∼1 milliarcsecond are possible, yielding distances good to 10% even at 100 pc, and allow astronomers to create an accurate three-dimensional map of the nearby . Much like early explorers’ discoveries created a more complete map of our Earth, accurate trigonometric parallaxes create a more detailed picture of the solar neighborhood, filling in the blank regions of nearby space. van Altena et al. (1995) compiled decades of ground-based work by dedicated as- trometrists in The General Catalogue of Trigonometric Stellar Parallaxes (also known as the “Yale Parallax Catalog”, hereafter, YPC). The YPC includes 15994 total trigonometric parallaxes for 8112 stars measured using photographic plates and CCDs published before the end of 1995. The Hipparcos space mission, launched in August 1989, provided ∼118,000 parallaxes (Perryman et al. 1997; ESA 1997, hereafter HIP), and was essentially complete –3– for stars with V = 7.3 − 9.0, while reaching to observe a few dozen stars even fainter than V ∼13. However, Hipparcos could not measure most nearby white, red, and brown dwarfs, which comprise more than 80% of the solar neighborhood population, and many of these systems have not yet had their distances measured via ground-based trigonometric parallax efforts. Here we present results for nearby systems observed during our Cerro Tololo Inter- american Observatory Parallax Investigation (CTIOPI). Twenty systems are new members of the Research Consortium on Nearby Stars (RECONS) sample, meeting the RECONS requirements that they have trigonometric parallaxes, πtrig, larger than 100 mas with errors less than 10 mas (see section 6.1). We provide a brief dossier for each new systems, including

VRIJHKs photometry and spectral types. We also provide improved πtrig for two additional important systems previously known to be within 10 pc, GJ 54 AB and GJ 1061, and for three systems that have reported πtrig larger than 100 mas, but which, in truth, lie beyond the 10 pc horizon. In total, we report 27 new trigonometric parallax measurements for 25 systems.

2. CTIOPI and Other Recent Trigonometric Parallax Efforts

Since the publication of the YPC and HIP results, only ∼250 ground-based parallaxes and a handful space-based parallaxes from HST (e.g., Benedict et al. 1999) have been pub- lished in the refereed literature. We provide a comprehensive tally of ground-based trigono- metric parallax measurements published for new RECONS systems in Table 1, since the 1995 cutoff date of the YPC.1 The transition of data acquisition techniques from photographic plates to CCDs and infrared arrays is nearly complete, although approximate parallaxes for the nearest objects can still be derived from large archival databases that include scanned plates (Deacon & Hambly 2001; Teegarden et al. 2003; Deacon et al. 2005). These parallaxes are useful but necessarily crude, given the limited number of epochs available and the coarse plate scales. The most recent change in parallax work is the advent of IR arrays (e.g., Vrba et al. 2004). Our CCD parallax effort, CTIOPI, began as an NOAO Surveys Program in 1999 August

1All parallaxes in Reid et al. (2000) were updated in Dahn et al. (2002). Two objects, TVLM 513-46546 (Tinney et al. 1995; Dahn et al. 2002) and 2MA 1047+2124, (Tinney et al. 2003; Vrba et al. 2004) are borderline objects — the Tinney parallaxes place the objects closer than 10 pc whereas the Dahn and Vrba USNO parallaxes place them beyond 10 pc. Weighted means of the two values for each object yield distances larger than 10 pc, so neither is included as a new 10 pc system in Table 1. –4– using both the 0.9 m and 1.5 m telescopes at CTIO, and has continued on the 0.9 m as part of the SMARTS (Small and Moderate Aperture Research Telescope System) Consortium be- ginning in 2003 February. The primary goals of CTIOPI are to discover and characterize any nearby objects that remain unidentified in the solar neighborhood — primarily red dwarfs, brown dwarfs, and white dwarfs. Although CTIOPI concentrates on southern hemisphere systems, we also target important nearby star candidates as far north as δ = +30. Most of the stellar systems observed during CTIOPI have been selected because available astro- metric data (high proper motions, crude parallaxes) or photometric/spectroscopic distance estimates indicate that they might be within 25 pc, the horizon of the Catalog of Nearby Stars (Gliese & Jahreiß 1991) and NStars (Henry et al. 2003) compendia. Objects possibly nearer than 10 pc are given highest priority. The parallaxes we measure typically have errors less than 3 mas, corresponding to distance errors of only 1–3% for systems within 10 pc. At the beginning of 2006, we continued to observe roughly two-thirds of the more than 400 systems targeted for parallax measurements during the 0.9 m program. Under SMARTS, we have expanded CTIOPI to carry out a program to search for low mass companions to nearby stars, called ASPENS (Astrometric Search for Planets Encircling Nearby Stars), in collaboration with David Koerner at Northern Arizona University. This paper is the fourth publication in The Solar Neighborhood series that includes CTIOPI parallaxes. Jao et al. (2005) presented 46 parallaxes for high systems observed at the 0.9 m. Costa et al. (2005) and Costa et al. (2006, AJ in press) presented 62 total parallaxes for fainter systems, including brown dwarf candidates, observed at the 1.5 m. In this paper, we concentrate on systems observed at the 0.9 m that were candidates for the 10 pc sample and present 27 total parallaxes.

3. Photometry

Because of the significant investment in observing time required to determine a trigono- metric parallax — roughly one integrated night over at least two for each target — it is prudent to obtain characterization photometry of nearby star candidates before placing them on the astrometric program. As shown in Henry et al. (2004), the combination of 2 VJ , RKC and IKC (hereafter, without subscripts) photometry and infrared J, H, and Ks photometry from 2MASS can be used in a suite of color-absolute magnitude relations to estimate distances of stars with 15% accuracy.

2 Subscript: J = Johnson, KC = Kron-Cousins. The central wavelengths for VJ , RKC and IKC are 5475A,˚ 6425A˚ and 8075A,˚ respectively. –5–

V RI photometry is reported in Table 2. Two names are often given, followed by the new optical V RI photometry and the number of nights on which observations were taken during CTIOPI. Photometry was acquired at the CTIO 0.9 m using the same filter and detector combination used for the astrometry frames (see section 5 for details of the instru- mental setup). All observations were taken between November 1999 and December 2005.3 Data were reduced via IRAF with typical bias subtraction and dome flat-fielding, using cal- ibration frames taken at the beginning of each night. Standard star fields from Graham (1982), Bessel (1990), and/or Landolt (1992) were observed several times each night to de- rive transformation equations and extinction curves. Apertures 14′′ in diameter were used to determine the stellar fluxes to match the apertures used by Landolt, except in cases when close sources needed to be deconvolved (LP 771-095 ABC) or excised (LHS 288, WT 460), in which case smaller apertures were used and aperture corrections were done. Further details about data reduction, transformation equations, etc., can be found in Jao et al. (2005). As discussed in Henry et al. (2004), representative total errors in the V RI photometry are ± 0.03 mag. Of the 81 V RI photometric values reported here, 69% have formal total 1σ errors of 0.03 or less. Only eight values have errors in excess of 0.05 mag: G 099-049, GJ 300, GJ 2130A, LHS 1610, and LHS 1723 at V with errors of 0.05 to 0.07 mag, and WT 460 with errors of 0.09, 0.06 and 0.05 at V RI, respectively, because it is crowded by nearby sources and required apertures of 5′′ or smaller and large aperture corrections. A detailed variability study for the parallax targets has been carried out by comparing the photometry of the target stars to the reference stars that set the astrometric grid for each field. Our methodology matches that of Honeycutt (1992), where details of the algorithm used can be found. Columns 7–10 of Table 2 list the filter used for parallax frames, the standard deviation of the target’s magnitude in that filter, the number of different nights the target was observed, and the total number of frames evaluated for variability. In general, stars with standard deviation values (relative to the reference stars) greater than 0.02 mag are obviously variable, those with values 0.01–0.02 mag are variable at a level of a few percent, and those with values less than 0.01 mag are “steady.” However, these cutoffs are not absolute because if the reference stars used to measure the flux modulations are faint, an artificially high standard deviation value may result. Therefore, a ‘v” is given after the value in Column 8 if variability has been detected reliably. In total, we identify six of the 27 targets to be certain variables, a rate of 22%. This is a lower than the 49% rate (21 of 43 red dwarfs) found to be variable by Weis (1994), primarily because our variability threshold

3The Tek #2 V RI filter set at CTIO has been used. However, the Tek #2 V filter cracked in March 2005 and was replaced by the very similar Tek #1 V filter. Reductions indicate no significant differences in either astrometric or photometric results from the two filters. –6– is roughly twice as high as that of Weis. Inspection of the photometric results indicates that many of the 20 new nearby systems are relatively bright, with eight systems having V < 13. Only one of these systems, LP 771-095 ABC, was observed by Hipparcos, but the close proximity of two sources, A and BC, resulted in a trigonometric parallax with high error, 92.97 ± 38.04 mas.

Infrared photometry in the JHKs system (rounded to the hundredth) has been extracted from 2MASS and is given in Columns 11–13 of Table 2. Because of the stars’ generally large

proper motions, each identification was confirmed by eye. The JHKs magnitude errors that include target, global, and systematic terms, i.e. the x−sigcom errors (where x is j, h, or k), are almost always less than 0.05 mag and are typically 0.02-0.03 mag. The only two exceptions are H band magnitude errors for LHS 337 (0.06 mag) and GJ 1207 (0.08 mag). Column 14 of Table 2 provides a distance estimate and error for each system, based on the suite of photometric distance relations in Henry et al. (2004). Column 15 lists the number of distance relations valid and used for each target, where the maximum is 12. These distances will be compared to the distances determined from the trigonometric parallaxes reported in section 6.2 to confirm or reveal new multiple systems, as discussed in section 7.2.

4. Spectroscopy

After the acquisition of optical and infrared photometry and the subsequent distance estimates, a final spectroscopic check is typically done to confirm the likely proximity of a candidate. Spectral types are useful for the separation of dwarfs, subdwarfs, and the occasional giant that slips through the photometric net. As part of a long-term program to characterize nearby star candidates, spectra were acquired between March 2002 and September 2004 on the CTIO 1.5 m and 4.0 m telescopes. On the 1.5 m telescope, observations were made using a 2′′.0 slit in the RC Spectrograph with grating #32 and order blocking filter OG570 to provide wavelength coverage from 6000 to 9500 A˚ and resolution of 8.6 A˚ on the Loral 1200x800 CCD. On the 4.0 m telescope, observations were made using a 2′′.0 slit in the RC Spectrograph with grating G181 and order blocking filter OG515 to provide wavelength coverage from 5000 to 10700 A˚ and resolution of 5.6 A˚ on the Loral 3K x 1K CCD. At least two sequential exposures were taken of each target to permit cosmic ray removal. Reductions were accomplished using standard methods in IRAF, and mild fringing, a result of the rear-illuminated CCD on the 4.0 m was removed with a tailored IDL routine. –7–

As shown in the last column of Table 2, all of the new nearby stars are red dwarfs, with types spanning M3.0V to M8.5V. What is somewhat surprising is that many of the new systems are not late-type M dwarfs — 16 of the 20 primaries in the new RECONS systems have spectral types earlier than M6.0V. Thus, there apparently remain many mid-type M dwarfs near the Sun that are not yet recognized as solar neighbors.

5. Astrometry Observations and Reductions

The 0.9 m telescope is equipped with a 2048 × 2048 Tektronix CCD camera with 0′′.401 pixel−1 plate scale that has remained on the telescope throughout CTIOPI. All observations of the parallax targets (“pi stars”) listed in Table 3 (coordinates in and equinox 2000.0) were made through V RI filters using the central quarter of the chip, yielding a 6.′8 square field of view. The same filter (Column 7 of Table 2 and Column 4 of Table 3) is used for all astrometric observations for a given pi star. The basic data reduction for the astrometry CCD frames includes overscan correction, bias subtraction and flat-fielding, using calibration frames taken at the beginning or end of each night. An extensive discussion of the data acquisition and reduction techniques is given in Jao et al. (2005). Briefly, 5–10 parallax frames are typically taken for a parallax field at each epoch to provide multiple measurements of pi and reference star centroids. Frames are usually taken within ±30 minutes of a pi star’s transit in order to minimize the corrections required for differential color refraction (DCR), with the goal of having some frames on both sides of the meridian on each night that a field is observed. All observations for a given field are made with a set of reference stars suitably positioned on the chip within a few pixels of their original locations. Good reference star configurations include 5–15 stars surrounding the pi star (the number of reference stars is given in Column 9 of Table 3). Typically, selected reference stars have at least 1000 counts at the peak and minimal proper motion and parallax (the ensemble of proper motion and parallax values for reference stars is set to zero during reduction). These reference stars are used as the grid against which the pi star astrometry is measured, to correct for any field translation or rotation caused by temporal shifts in the telescope/camera combination, and as comparison stars for variability studies. In addition, V RI photometry is obtained for the reference stars and used to make DCR corrections and to estimate distances to each star photometrically to permit the conversion from relative to absolute parallax. To decouple parallactic and proper motions in the final astrometric solution, observa- tions are typically taken over at least two years, including four high parallax factor seasons. Columns 5–8 of Table 3 list the number of seasons, total number of frames, epochs of ob- –8– servation, and length of time for each pi star frame series. As in previous papers, in the seasons column, “c” indicates a continuous set of observations throughout the seasons and “s” indicates scattered observations, meaning that the target was observed on only one night during one or more seasons (or was missed entirely during a season); continuous observations are superior to scattered observations. The set of observations for a given pi star usually includes at least 20 evening and 20 morning frames. However, because the pi stars in this paper were likely to be within 10 pc, they were given high priority status throughout their tenures on the observing list; thus, the total number of frames given for each target often far exceeds the typical 40 total frames for a pi star.

6. Astrometry Results

The trigonometric parallax results are given in Table 3, including both relative (Col- umn 10) and absolute (Column 12) parallaxes, as well as the the correction between the two (Column 11). The corrections are generally less than ∼2 mas, so systematics in the cor- rections should not significantly affect the final results. Proper motions and position angles of the proper motions are given in Columns 13 and 14, and the derived tangential velocity is given in Column 15. The final column indicates whether previous trigonometric values were available, with notes following the table. After outlining the definition of a RECONS system, we discuss each system individually.

6.1. Definition of RECONS Systems

Stars targeted in this study include potential new 10 pc sample members with no trigono- metric parallaxes and stars supposedly within 10 pc with large trigonometric parallax errors. The second category of targets allows us the opportunity to evaluate the threshold of reli- ability for available trigonometric parallaxes. As supported by the results described below, we formally define the parameters for inclusion in the RECONS 10 pc sample as: A system comprised of stars, brown dwarfs, and/or planets (and any associated surrounding material) for which a trigonometric parallax of at least 100 mas has been determined, and for which the parallax has a formal error of 10 mas or less. The advantages of this definition are (1) it assigns a clear “in” or “out” status for each system, rather than using probabilities for systems, in particular those near the 10 pc border; (2) it establishes a uniform 10 mas error cutoff for every system in the sample; and (3) it provides a limit on the “worst case” for any system because the distance is known to –9– better than 10% in all cases. This final benchmark is superior to current photometric or spectroscopic distance estimates that, when done correctly, are not better than 15% (Henry et al. 2004) for red dwarfs, the dominant component of the nearby star population. In truth, we do not consider a trigonometric parallax “definitive” unless its error is less than 3 mas, and this value may change when large numbers of stars are measured more accurately via future efforts. However, we include systems with parallax errors of 3–10 in the RECONS sample to bolster the statistics until we can acquire improved data, as is our goal during CTIOPI. We do not include systems with parallax errors in excess of 10 mas, because, as shown in section 6.4, many systems with errors this large do not prove to be within 10 pc. One awkward aspect of the adopted definition is that it excludes systems with very large parallaxes and large parallax errors until accurate data are in hand. Examples include stars with approximate parallaxes measured using a few photographic plates, such as SO

0253+1652 and SCR 1845-6357 AB, and systems with πtrig from more extensive plate series having errors slightly in excess of 10 mas, such as G 041-014 ABC and LHS 288, both of which were reported to be within five (see section 6.2). LHS 288 is a particularly subtle example of applying the RECONS criteria for membership. Ianna & Bessell (1986) determined a parallax of 221 ± 8 mas, but the YPC, which carries out a systematic assess- ment of parallaxes and errors, gives an error of 11.3 mas for LHS 288’s parallax. Thus, this star did not meet the formal definition for a RECONS member because we used the YPC for sample definition. These “nearly certain” members were generally observed early and intensely in CTIOPI to bring them formally into the sample.

6.2. New RECONS Systems

Here we report 20 new RECONS stellar systems based on the trigonometric parallaxes given in the top portion of Table 3. Previously, none of these systems had trigonometric parallaxes with errors less than 10 mas, and 12 had no trigonometric parallaxes of any kind. Of the 20 new systems, five have proper motions in excess of 1′′.0/, 11 have motions between 1′′.0/year and 0′′.5/year, and four have motions less than 0′′.5/year. These numbers indicate that there are likely to be large numbers of stars within 10 pc of the Sun that have low, but not negligible, proper motions. Many of these systems have been recovered through our extensive efforts to estimate reliable distances for stars included in a large collection of potentially nearby objects. The collection includes various types of astrometric, photometric, and spectroscopic data that – 10 – have been converted to standard systems, evaluated for reliability, and then used to pinpoint systems potentially nearer than 10 pc. Primary sources include (1) the valuable Catalog of Nearby Stars (Gliese & Jahreiß 1991, hereafter CNS), which includes many stars with photometric parallaxes, (2) the Luyten Half Second Catalogue (Luyten 1979), which includes 3602 objects with proper motions in excess of 0′′.5 yr−1, and which still accounts for 87% of all known such objects as of this writing (see Subasavage et al. 2005b), and (3) the monumental work of Weis and collaborators, who provided accurate optical photometry and computed photometric parallaxes for thousands of stars (Weis 1984, 1986, 1987, 1988; Booth et al. 1988; Weis 1991a,b, 1993, 1996). More recently, large scale work by Reid and collaborators (Reid & Cruz 2002; Reid et al. 2002; Cruz & Reid 2002; Reid et al. 2003a; Cruz et al. 2003; Reid 2003; Reid et al. 2003b, 2004) has identified additional new systems from proper motion compendia, and has reidentified and confirmed many of the systems noted by others. In addition, our own new recent proper motion survey, known as the SuperCOSMOS-RECONS (SCR) effort (Hambly et al. 2004; Henry et al. 2004; Subasavage et al. 2005a,b) has already provided three new systems described in this paper. The initial dates for the astrometric series listed in Table 3 indicate when the targets were first observed, illustrating that many targets were placed on the parallax program before most of the more recent distance estimating efforts. Here we provide a brief dossier for each new RECONS sample system. The notes to Table 3 list previously available trigonometric parallaxes, which are also given here in the text along side photometric or spectroscopic distances estimates. It is also important to note that the trigonometric parallaxes given in the 1991 edition of the CNS are from a preliminary version of the YPC. Thus, they are of the same origin as the definitive YPC, published in 1995, and are therefore not listed explicitly here. The final YPC parallaxes often have somewhat larger errors determined systematically during construction of the catalog, and errors in excess of 10 mas exclude some stars from the RECONS sample until the present study, even though their CNS errors were less than 10 mas. LHS 1302 was reported to have photometric distances of 10.0 ± 1.8 pc in CNS and 10.2 ± 1.3 by Reid & Cruz (2002). The distance is 9.92 ± 0.19 pc, consistent with previous estimates and near the horizon of the RECONS sample. There is a hint of a perturbation in the RA residuals that may be confirmed or refuted during via ASPENS. The star is variable at a level of 0.05 mag in the R band during 26 nights of observation, and flared by 0.13 mag in R on UT 2004 November 22 during a sequence of ten frames spanning 37 minutes. APMPM J0237−5928 is an X-ray source reported by Scholz et al. (1999) to have a photometric distance between 11 and 14.5 pc based on a spectral type of M5V. The distance is 9.64 ± 0.10 pc. The star is mildly variable, with a standard deviation of 0.013 mag in the – 11 –

R band during 27 nights of observation, with a full variability range of 0.06 mag. SO 0253+1652 was reported by Teegarden et al. (2003), to have a crude trigonometric parallax of 410 ± 90 mas (2.4 ± 0.6 pc) that was described as a “lower limit on the true parallax.” Based on this value, claims were made that it was the third nearest star system, was an extremely metal poor subdwarf, or had a radius only 60% that of a comparable star, GJ 1111. However, their photometric estimate (3.6 ± 0.4 pc) and our group’s estimate (3.7 ± 0.6 pc, Henry et al. 2004) are better matches to the actual distance of 3.84 ± 0.04 pc determined here. Rather than being an exotic object, SO 0253+1652 is a normal

with MV = 17.22 and spectral type M7.0V, and is the 23rd nearest system to the Sun. LP 771−095 ABC is a triple system (A is LP 771-095 and the close pair BC is LP 771-096) with separations of 7′′.22 at PA = 315◦ for A–BC (Jao et al. 2003) and 1′′.30 at PA = 138◦ for B–C (this paper), with no change in position angle noted during the six years of our observations. A and BC have separate photometry, with A brighter than BC at V and R, but fainter at I (see Table 2). The magnitude differences between B and C are ∆V =0.86 ± 0.09, ∆R =0.75 ± 0.03, and ∆I =0.66 ± 0.07. The CTIOPI reference field is faint and sparse, and as few as four reference stars were used to boost the frame count. Nevertheless, we measure consistent parallaxes for A and BC, although the parallax error for BC (4.99 mas) is large because of elongated images due to its duplicity that result in poorly determined centroids. The weighted mean of the two parallaxes is 144.68 ± 2.52 mas (6.91 ± 0.12 pc), a factor of 15 improvement in the 92.97 ± 38.04 mas (10.8 ± 5.3 pc) value measured by Hipparcos. The system was reported to have photometric distances of 7.6 ± 1.8 pc (for A and BC) in CNS, 7.8 pc in Weis (1991b), and 10.3 pc for A and 12.1 pc for BC by Reid et al. (2004). After deconvolving BC, the latter authors estimate a distance of 10.2 pc. The system is actually closer than all photometric estimates. On UT 1999 August 22, the BC pair showed a 0.30 mag decrease in brightness. No other night shows such a decrease, so we cannot confirm that this is an eclipsing event, but note it for future monitoring. Neither the A component, nor any of the reference stars show this anomaly. The standard deviations in variability for both A and BC are artificially high because the reference stars are faint — neither source is obviously variable. LHS 1610 has a trigonometric parallax in YPC of 70.0 ± 13.8 mas (14.3 ± 2.9 pc). The parallax value and its large error both exclude the star from the RECONS sample. LHS 1610 was reported to have photometric distances of 9.6 ± 1.4 pc in CNS and 10.5 ± 1.2 pc by Reid & Cruz (2002). The distance is 9.85 ± 0.20 pc, consistent with the photometric estimates and near the horizon of the RECONS sample. – 12 –

The star is obviously variable, with a standard deviation of 0.024 mag in the V band during 23 nights of observation. Parceling the data into individual nights shows that the star varies by 0.08 mag, possibly regularly. Continuing observations as part of ASPENS will provide additional epochs of photometry to further investigate the periodicity. LHS 1723 was reported to have photometric distances of 6.1 ± 1.0 pc in CNS, 9.2 pc by Henry et al. (1994), 9 pc by Patterson et al. (1998), and 5.7 ± 0.5 pc by Reid et al. (2002). The distance is 5.32 ± 0.04 pc. The star is obviously variable, with a standard deviation of 0.020 mag in the V band during 27 nights of observation, with a full variability range of 0.09 mag. G 099−049 has a trigonometric parallax in YPC of 186.2 ± 10.1 mas (5.4 ± 0.3 pc). The large error excludes the star from the RECONS sample until the present study. The formal correction from relative to absolute parallax is 4.50 ± 0.82 mas because the reference stars are red. The field is in Orion, so the redness is artificial, thereby causing the reference stars to appear much nearer than they are, with photometric parallaxes of 3–17 mas. The reference stars’ trigonometric parallaxes and proper motions are well-behaved, being less than 1 mas and 10 mas/yr, respectively. We have therefore adopted a reasonable “mean” correction to absolute parallax of 1.50 ± 0.50 mas (assuming a large error) instead of the erroneous formal correction. The star exhibits a possible perturbation of the photocenter at a level of ∼20 mas in the RA residuals; further investigation during ASPENS will help confirm or refute the perturbation. The source 6′′ to the NW in late 2005 is a background source, and may be affecting the centroids. Our derived proper motion, 0′′.313/yr, and direction of the proper motion, 98◦, are significantly different than given in CNS, 0′′.241/yr at 108◦. L´epine & Shara (2005) report a proper motion of 0′′.314/yr at 98◦, consistent with our value. SCR 0630−7643 AB was first reported in Henry et al. (2004) to be a potential nearby binary of type M6.0VJ with an estimated photometric distance of 7.0 ± 1.2 pc, assuming magnitude differences in VRIJHK of 0.25 mag. Subasavage et al. (2005a) estimated a distance of 6.9 pc based on photographic plate and 2MASS magnitudes. The separation of AB is 0′′.90 at PA = 34◦, with no change in position angle noted during the two year period of the observations. The magnitude differences between A and B are ∆V =0.20 ± 0.03, ∆R = 0.23 ± 0.06, and ∆I = 0.21 ± 0.10, thereby confirming that the two stars are similar in brightness and color. Frames with seeing better than 1′′.2 were removed from the reduction because the images are elongated. The trigonometric parallax reported here corresponds to a distance of 8.76 ± 0.14 pc, further than previous estimates. The projected separation is 7.9 AU, which implies an orbital period of nearly 50 yr for two – 13 –

stars with masses of 0.10 M⊙ each. G 089−032 AB was reported to be a single star with photometric distances of 5.8 pc by Weis (1986), 6.2 ± 1.0 pc in CNS, and 8.1 pc by Henry et al. (1994). The distance is 8.58 ± 0.07 pc — the distance estimates were generally too low because the system is a close double. The system was first reported to be a binary with separation 0′′.7 and of nearly equal brightness at K by Henry et al. (1999). During CTIOPI, the resulting FWHM measure- ments are typically > 1′′.5 because the unresolved B component stretches the images. The components are too close together for any clear measurements of magnitude differences in CTIOPI frames, but we can estimate the position angle of the companion to be ∼300◦, with no obvious change in the six years of coverage. At the measured distance, the projected separation is 6.0 AU, and implies an orbital period of ∼30 years, assuming masses of 0.13 M⊙ for each component. No perturbation can yet be identified in the RA and DEC residuals, consistent with the long orbital period. GJ 300 has a trigonometric parallax in YPC of 169.9 ± 15.0 mas (5.9 ± 0.5 pc). The large error excludes the star from the RECONS sample until the present study. Reid et al. (2004) reported a photometric distance of 6.3 pc. We measure a distance of 7.96 ± 0.06 pc, which is inconsistent with the previous trigonometric measurement and indicates that the uncertainty was optimistic. The star is mildly variable, with a standard deviation of 0.016 mag in the V band during 31 nights of observation, with a full variability range of 0.06 mag. G 041−014 ABC is a triple system in which A and B form a close spectroscopic binary, and C is more distant. Reid & Gizis (1997) report a separation of 1 AU for AB, while Delfosse et al. (1999) report an orbital period of 7.6 days. For component masses of 0.19 and 0.17 M⊙ for A and B (estimated after deconvolving the C component, then using the ratio of the spectroscopic K1 and K2 values and assigning masses via the mass- relation of Henry et al. (1999)), the separation would be only 0.05 AU. Nonetheless, the system is photometrically steady, with a standard deviation of only 0.009 mag in the V band during 22 nights of observation, indicating no obvious interaction between the components. Delfosse et al. (1999) imaged component C at a separation of 0′′.62 at PA ∼ 90◦ in 1997. The stars cannot be resolved in CTIOPI frames, so the photometry is for the combined light of all three stars. In fact, no elongation is seen even in frames with images having FWHM of 1′′.0. At the measured distance, the projected separation of AB–C is 4.2 AU, implying an orbital period of 11.5 years, assuming an additional mass of 0.17 M⊙ for component C. A hint of a perturbation at a level of 10 mas over six years can be seen in the RA residuals, but much more data are needed to confirm it. This perturbation is consistent with an orbital – 14 – period that may be much longer than that derived from the projected separation. The system was reported to have photometric distances of 4.5 ± 0.7 pc in CNS, 4.6 pc by Weis (1991b), and 5.3 pc by Henry et al. (1994), all of whom assumed it to be a single star. The distance is 6.77 ± 0.09 pc — the distance estimates were too low because the system is a close triple. LHS 2090 was reported to have photometric distances of 6.0 ± 1.1 pc by Scholz et al. (2001), 5.2 ± 1.0 by Reid & Cruz (2002), and 5.7 ± 0.9 pc by Henry et al. (2004). The distance is 6.37 ± 0.11 pc, consistent with previous estimates. LHS 2206 was reported to have a photometric distance of 10.2 ± 1.6 pc in CNS. The distance is 9.23 ± 0.20 pc, consistent with the previous estimate, and firmly placing the star in the RECONS sample. LHS 288 is a well-known nearby star in a very crowded field with a trigonometric parallax in YPC of 222.5 ± 11.3 mas (4.5 ± 0.2 pc). The large error excludes the star from the RECONS sample until the present study. However, the original result by Ianna & Bessell (1986), 221 ± 8 mas (4.5 ± 0.2 pc), has an error small enough to include the star in the RECONS sample. Fortunately, Ianna is an author on this paper as well as the original parallax paper, so can be credited with the discovery of LHS 288 as a RECONS system, whether the date is 1986 or at time of this paper. Henry et al. (2004) reported the star to have a photometric distance of 6.9 ± 1.7 pc. We measure a distance of 4.79 ± 0.06 pc, consistent with previous measurements. SCR 1138−7721 was first reported in Hambly et al. (2004) to be a potential nearby star with an estimated photometric distance based on plate and 2MASS magnitudes of 8.8 ± 2.7 pc (when both internal and external errors are included). Henry et al. (2004) provided a distance estimate of 9.4 ± 1.7 pc using CCD and 2MASS magnitudes. The distance is 8.18 ± 0.20 pc, consistent with the previous estimates. LHS 337 was reported to have photometric distances of 7.2 ± 1.5 pc in CNS and 9 pc by Patterson et al. (1998). The distance is 6.38 ± 0.08 pc. WT 460 was reported to have a photometric distance of 11 pc by Patterson et al. (1998). The distance is 9.31 ± 0.13 pc, firmly placing the star in the RECONS sample. The star is corrupted by two sources within 6′′ in the NE quadrant, neither of which is a common proper motion companion. Astrometric residuals with RMS values of 24 mas in RA and 32 mas in DEC are roughly three times typical residuals, but the large number of frames and strong reference star configuration allow an accurate parallax to be determined. This star will not be followed in ASPENS because of the corrupting background sources. – 15 –

GJ 1207 has a trigonometric parallax in YPC of 104.4 ± 13.6 mas (9.6 ± 1.3 pc). The large error excludes the star from the RECONS sample until the present study. We measure a distance of 8.67 ± 0.11 pc, which is consistent with the previous measurement. GJ 1207 has the largest photometric standard deviation — 0.263 mag in the V band during 27 nights of observation — of any star reported here, primarily because of one extreme flare event. On UT 2002 June 17, the star’s brightness was normal relative to reference stars in the first frame, then increased by a factor of 4.6 (1.7 mag) by the second frame, and subsequently faded to a factor of 2.5 (1.0 mag) higher than normal by the sixth frame. The entire sequence spanned only 15 minutes. SCR 1845−6357 AB was first reported in Hambly et al. (2004) to be a very red, potential nearby star with an estimated photometric distance based on plate and 2MASS magnitudes of 3.5 ± 1.2 pc (when both internal and external errors are included). A spectral type of M8.5V and distance estimate of 4.6 ± 0.8 pc were presented using CCD and 2MASS magnitudes in Henry et al. (2004). Deacon et al. (2005) measured a preliminary trigonometric parallax, 282 ± 23 mas (3.5 ± 0.3 pc), based on eight SuperCOSMOS photographic plates, illustrating that reasonable parallaxes can be derived with only a few archival plates. We measure a parallax of 259.45 ± 1.11 mas (3.85 ± 0.02 pc), consistent with all previous estimates. This distance makes SCR 1845-6357 AB the 24th nearest system to the Sun. Biller et al. (2006) report a T dwarf companion separated by 1′′.17 at position angle 170◦ in 2005. We find a very low error in the parallax because of intense coverage and a rich, well-balanced set of reference stars. We do not yet see any curvature in the the astrometric centroids due to the companion, but the residuals do show a larger scatter than for most stars, possibly due to the pull of the companion. Assuming masses of 0.10 and 0.05 M⊙ and a 4 AU separation, we derive a 21-year orbital period. The photocentric semimajor axis is predicted to be ∼350 mas, corresponding to a perturbation in the photocenter of ∼34 mas/year, although the rate depends on the true size, shape, and orientation of the orbit on the sky. Continued observations are planned to improve and monitor the position of the center of mass of the system. LHS 3746 was reported to have photometric distances of 10.0 ± 2.7 pc in CNS and 10 pc by Patterson et al. (1998). The distance is 7.45 ± 0.07 pc. The relative photometry shows a long-term trend possibly indicative of a stellar activity cycle.

6.3. Known RECONS Systems

GJ 54 AB is a fast-orbiting pair of red dwarfs that Rodgers & Eggen (1974) reported – 16 – as a “Suspected very close binary in the 40-inch reflector. The components are assumed to be of equal magnitude.” No separation, position angle, or epoch were given, and the separation must have been near the diffraction limit of the Siding Spring Observatory 40- inch. Golimowski et al. (2004) provide details of the system, including clear resolution of the pair using HST-NICMOS. The system has been followed since 2000 in our continuing HST- FGS effort to measure accurate masses for low mass stars (Henry 2004). The HST − FGS measurements indicate that ∆V = 1.04, and a preliminary orbital analysis yields an orbital period of ∼1.1 years and semimajor axis of 0.9 AU. As can be seen in Figure 1, the perturbation affects the astrometric residuals in both RA and DEC (9.1 and 13.6 mas RMS, respectively) after solving for proper motion and parallax. For comparison, the “well-behaved” GJ 300 RA and DEC residuals (3.6 mas RMS in each axis) are shown. For GJ 54 AB, beating of the 1.0 year parallax motion and 1.1 year orbital motion makes deconvolution of the two motions problematic, and a high parallax error (3.38 mas) results (and is not helped by a faint set of reference stars, which also cause the variability measurement to be artificially high). In fact, this is nearly a worst-case scenario because of the confluence of the two perturbations on the star’s straight-line proper motion path, coupled with the small semimajor axis of the photocenter’s orbital motion, only 16 mas. The system has trigonometric parallax measurements listed in YPC and HIP of 120.5 ± 10.1 mas (8.30 ± 0.70 pc) and 122.86 ± 7.53 mas (8.14 ± 0.50 pc), respectively. Our data yield a better parallax, but the weighted mean of the three independent parallax measurements, 136.62 ± 2.95 mas (7.32 ± 0.16 pc), still has a relatively high error. The system will be observed in ASPENS to improve the parallax, monitor the motion of the photocenter, and facilitate the derivation of accurate masses. GJ 1061 was reported to have a photometric distance of 4.3 ± 1.1 pc in CNS. Henry et al. (1997) discovered the star to be the 20th nearest stellar system with a trigonometric parallax of 273.4 ± 5.2 mas (3.66 ± 0.07 pc) from photographic plates. Three other stars of similar nearby pedigree reported in that work turned out to be distant giants, providing classic illustrations of how photometry and spectroscopy can be used to fine-tune astrometry observing lists, before spending observing time on stars that will have negligible parallax. Here we provide a CCD parallax of 271.92 ± 1.34 (3.68 ± 0.02 pc) for GJ 1061, which has an error four times smaller than the previous value and is completely consistent. GJ 1061’s remains the 20th nearest system.

6.4. Not RECONS Systems

GJ 1001 ABC is a triple system with separations of 18′′.2 at PA = 259◦ for A–BC and – 17 – a changing separation always less than 0′′.1 for B-C (Golimowski et al. 2004). The initial L dwarf companion was reported by the EROS Collaboration (1999), and was resolved by Golimowski et al. (2004) using HST-NICMOS into two nearly equal magnitude L dwarfs.

The BC pair is evident in the parallax frames as a very faint source, but no πtrig measurement was attempted. GJ 1001 A has a trigonometric parallax in YPC of 104.7 ± 11.4 mas (9.6 ± 1.1 pc). The distance determined here is 13.01 ± 0.67 pc, clearly pushing the system beyond the horizon of the RECONS sample. So far, the improvement in the parallax error is only a factor of three because of a set of faint reference stars and poor observational coverage. The primary will remain a CTIOPI target because an accurate parallax for the system offers the opportunity to determine crucial masses for the brown dwarfs, B and C, which continuing observations show have an orbital period of ∼4 years (Golimowski et al. in prep). GJ 633 has a trigonometric parallax in YPC of 104.0 ± 13.7 mas (9.6 ± 1.3 pc). The distance determined here is 22.48 ± 0.93 pc, well beyond the 10 pc limit of the RECONS sample. A photocenter perturbation of ∼60 mas is seen in the RA residuals, which may be real or could be due to a background star 7′′ to the E that corrupts centroid measurements. The distance derived using VRIJHK photometry and the suite of relations in Henry et al. (2004) is 19.5 ± 3.0 pc, which is consistent with the trigonometric distance. The target will remain on the program to determine the veracity of the perturbation and the existence of a possible companion that would contribute minimal light to the system. GJ 2130 ABC is a triple system with a separation of 21′′.23 at PA = 88◦ for A–BC (Jao et al. 2003, incorrectly noted as AC and B). BC is reported to be a double-lined spectroscopic binary by Delfosse et al. (1999). The duplicity of BC is not evident in frames with images having FWHM of 1′′.0. Gliese & Jahreiß (1979) provided distance estimates of 14.5 pc for A and 19.1 pc for B (C was not known at the time). Reid et al. (2004) provided distances of 13.5 pc for A and 6.2 pc for BC (adopted from HIP). The distance estimates for the presumed single star, A, are good matches to our consistent measurements of parallaxes for A and BC, corresponding to 14.00 ± 0.52 pc and 14.24 ± 0.42 pc, respectively. The weighted mean of the two parallaxes for the system is 70.69 ± 1.62 mas (14.15 ± 0.32 pc), which is a factor of 14 improvement in the value measured by Hipparcos, 161.77 ± 11.29 mas (6.2 ± 0.4 pc), and a factor of five improvement in the value of 68.5 ± 7.9 mas (14.6 ± 1.7 pc) found by Fabricius & Makarov (2000), who used Hipparcos transit data to improve the result. – 18 –

7. Discussion

7.1. The Solar Neighborhood Population

After the publication of the YPC in 1995 and Hipparcos results in 1997, there were 215 systems in the RECONS sample. Between the time of those publications and the first parallax results from CTIOPI (2005), only eight new systems had reliable trigonometric parallaxes measured that placed them within 10 pc (see Table 1), and the nearest, GJ 1061, was by our group (Henry et al. 1997). CTIOPI has already reported eight additional new 10 pc systems meeting the RECONS criteria — five in Jao et al. (2005), two in Costa et al. (2005), and one in Costa et al. (2006, AJ in press). In this paper, we add an additional 20 systems to the sample of stars within 10 pc with high quality parallaxes. Counting additions in this paper, since the publication of YPC and HIP results, the total number of new RECONS systems stands at 36. The 34 systems added since 2000 constitute a 16% increase in the number of systems within 10 pc in six years.

Figure 2 shows an HR diagram of the RECONS sample, plotting MV vs. V − K for all systems with parallaxes meeting the RECONS criteria (see Section 6.1). All multiples have been deconvolved into individual components as well as can currently be done at both the V and K bandpasses. Small circles indicate objects known to be within 10 pc after publication of the YPC and HIP catalogs, as well as new discoveries by other groups. Solid points represent RECONS discoveries. Labels are given for prominent or unusual stars. Triangles mark the four known subd- warfs within 10 pc. AU Mic and AT Mic AB (GJ 803 and GJ 799 AB, respectively) form a wide triple system that is apparently quite young. The system appears to be comoving with the β Pic group and has an age of only ∼20 Myr. The position of all three stars above the main sequence supports the hypothesis that the system is young, and the stars have not yet descended onto the main sequence. Further support came recently, when AU Mic was found to be the nearest (9.9 pc) star with a dust disk directly observed at optical and near-infrared wavelengths (Kalas et al. 2004). The three known L dwarfs within 10 pc — 2MA 0036+1821, 2MA 1507-1627, and DEN 0255-4700 — are shown, but none of the eight known T dwarfs or four extrasolar planets are plotted. All of the systems reported in this paper are mid- to late-type M dwarfs (spectral types listed in Table 2). The three brightest new stars are labeled — LP 771-095 A, LHS 3746, and GJ 1207 — as well as many of the fainter discoveries from the RECONS group and others. Note in particular the MV of 24.4 for DEN 0255-4700, for which a parallax and photometry were determined during our CTIOPI 1.5m program (Costa et al. 2006, AJ in press). We believe that this is the faintest object outside the Solar System for which MV – 19 – has been determined. The dominance of the nearby stellar population by red dwarfs is obvious from Figure 2. For objects in the refereed literature, including this paper, the complete count of current objects meeting the RECONS criteria is: 18 white dwarfs, 4 A types, 6F,21G,44K,240M, 3 L, 8 T, and 4 extrasolar planet candidates. Counts include the Sun and the adoption of a

dividing line between spectral types K and M at MV = 9.00. Thus, of the 348 total objects, the 240 M dwarfs comprise 69% of the nearby population. Elimination of the presumably non-stellar L, T, and extrasolar planet portions brings the fraction to 72% of all stars, and this fraction will continue to grow as more nearby red dwarfs are revealed.

7.2. Hidden Multiples

To confirm or reveal unresolved multiple systems, we make a comparison between the accurate trigonometric distances presented here and in previous CTIOPI papers and the

photometric distances estimated via the suite of MK -color relations described in Henry et al. (2004). This technique provides distance estimates good to 15%, measured by running RECONS stars with accurate trigonometric parallaxes back through the suite of relations, under the assumption that the stars are single. Photometric distance estimates for the systems discussed in detail in this paper are given in column 14 of Table 2, and the number of relations used for each target (maximum of 12) is given in column 15. Figure 3 compares the photometric distances with trigonometric distances for objects within 20 pc, as reported by CTIOPI in the four papers with parallaxes to date. The distance cutoff of 20 pc has been selected because that is the distance limit of our 20-20-20 sample (within 20 pc, orbital periods less than 20 years, at least one component with a mass less than 20% of the Sun’s; for details see Henry et al. (1999)), which includes red dwarf binaries observed to measure high quality masses. Presumed single main sequence stars are shown as open circles and known close multiples with combined photometry are shown with filled squares. The solid line represents identical photometric and trigonometric distances. Dotted lines trace distances differing by 20% — stars above the outlined region are often subdwarfs; stars below are often unresolved multiples because they are predicted photometrically to be closer than they actually are. In two cases, SCR 1845-6357 AB and LP 771-095 BC, the multiples are within the 20% region, as is expected because of the relatively large magnitude differences between the components. Additional close multiples in which secondaries contribute minimal light to the system would not be revealed with this method. From left to right (increasing – 20 – trigonometric distance), the known close multiples recovered below the 20% region are G 041-014 ABC, GJ 54 AB, GJ 2005 ABC, G 089-032 AB, SCR 0630-7643 AB, and GJ 2130 BC. The three best candidates for previously unnoticed multiples are shown as heavy open squares — from left to right, GJ 555, ER 2, and LTT 6933. GJ 555 is a marginal case, while the latter two are almost certainly close multiples with small flux ratios.

7.3. The Big Picture

To date, we have observed more than 400 stars for parallax during CTIOPI. This paper is the fifth in our series that provides accurate parallaxes for nearby stellar systems — one from Ianna’s plate parallax effort at Siding Spring Observatory and four from the ongoing CCD effort at CTIO. With the 20 new systems described here, we have now reported high quality parallaxes for a total of 29 new RECONS systems. We anticipate that future papers will yield additional nearby systems because the pool of stars nearer than 10 pc without accurate parallaxes is by no means drained. We emphasize that the stellar companions discussed here are not at CTIOPI’s detection threshold — the ASPENS program will search for substellar and possibly planetary mass companions. Red and white dwarf systems within 10 pc and south of δ = 0, including all but one (WT 460) of the new RECONS members discussed here, are being observed intensely to reveal any possible long term astrometric perturbations. Initial results imply that we can

detect companions with masses as low as 10 MJup for stars with good observational coverage and strong reference star sets. Each new system found within 10 pc is a member of the fundamental sample of ob- jects that is used to: (1) determine accurate luminosity and mass functions in the solar neighborhood, (2) evaluate the multiplicity of stellar populations, and (3) calculate the to- tal amount of mass found in stars and substellar objects. Because of their proximity, each system is a promising target for detailed astrophysical studies such as astroseismology, vari- ability investigations, direct radii measurements, and metal abundance evaluations. Aside from astrophysical efforts, these systems have great astrobiological potential because of their proximity. They are prime targets to be scrutinized as possible planetary hosts, and any planets found can be explored as nearby harbors of life. – 21 –

8. Acknowledgments

We would like to thank Georgia State University students who have assisted in many facets of the RECONS effort to discover nearby stars, including Jacob Bean, Misty Brown, Charlie Finch, Stephanie Ramos, and Jennifer Winters. We also gratefully acknowledge assistance in the early stages of the CTIOPI effort from Claudio Anguita, Rafael Pujals, Maria Teresa Ruiz and Pat Seitzer. Nigel Hambly has been vital to our SuperCOSMOS search, which revealed the three SCR stars here for which we report parallaxes. We also thank Hartmut Jahreiß, a referee without peer for this paper, for his review. Without the extensive observing support of Alberto Miranda, Edgardo Cosgrove, Arturo Gomez and the staff at CTIO, the results presented here would not be possible. We are deeply indebted to NOAO for providing us a long term observing program at CTIO via the NOAO Surveys Project. We also thank the continuing support of the members of the SMARTS Consortium without whom many of the astrometric series reported would not have been completed. The early phase of CTIOPI was supported by the NASA/NSF Nearby Star (NStars) Project through NASA Ames Research Center. The RECONS team at Georgia State Uni- versity (GSU) is supported by NASA’s Space Interferometry Mission, the National Science Foundation (grant AST-0507711), and GSU. Ianna’s efforts at the University of Virginia have been supported by the National Science Foundation (grant AST-9820711). EC and RAM acknowledge support by the Fondo Nacional de Investigaci´on Cient´ifica y Tecnol´ogica (proyecto Fondecyt No. 1010137), and by the Chilean Centro de Astrof´isica FONDAP (No. 15010003). This work has used data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center at California Institute of Technology funded by NASA and NSF.

REFERENCES

Benedict, G. F., et al. 1999, AJ, 118, 1086

Bessel, M. S. 1990, A&AS, 83, 357

Biller, B. A., Kasper, M., Close, L. M., Brandner, W., & Kellner, S. 2006, ApJ, 641, L141

Booth, J., Caruso, J., & Weis, E. W. 1988, PASP, 100, 749

Costa, E., M´endez, R. A., Jao, W.-C., Henry, T. J., Subasavage, J. P., Brown, M. A., Ianna, P. A., & Bartlett, J. 2005, AJ, 130, 337 – 22 –

Costa, E., M´endez, R. A., Jao, W.-C., Henry, T. J., Subasavage, J. P., & Ianna, P. A. 2006, AJ, in press

Cruz, K. L., & Reid, I. N. 2002, AJ, 123, 2828

Cruz, K. L., Reid, I. N., Liebert, J., Kirkpatrick, J. D., & Lowrance, P. J. 2003, AJ, 126, 2421

Dahn, C. C., et al. 2002, AJ, 124, 1170

Deacon, N. R., & Hambly, N. C. 2001, A&A, 380, 148

Deacon, N. R., Hambly, N. C., Henry, T. J., Subasavage, J. P., Brown, M. A., & Jao, W.-C. 2005, AJ, 129, 409

Delfosse, X., Forveille, T., Beuzit, J.-L., Udry, S., Mayor, M., & Perrier, C. 1999, A&A, 344, 897

EROS Collaboration, et al. 1999, A&A, 351, L5

ESA, 1. 1997, VizieR Online Data Catalog, 1239, 0

Fabricius, C., & Makarov, V. V. 2000, A&AS, 144, 45

Gliese, W., & Jahreiß, H. 1979, A&AS, 38, 423

Gliese, W., & Jahreiß, H. 1991, On: The Astronomical Data Center CD-ROM: Selected As- tronomical Catalogs, Vol. I; L.E. Brotzmann, S.E. Gesser (eds.), NASA/Astronomical Data Center, Goddard Space Flight Center, Greenbelt, MD

Golimowski, D. A., et al. 2004, AJ, 128, 1733

Graham, J. A. 1982, PASP, 94, 244

Hambly, N. C., Henry, T. J., Subasavage, J. P., Brown, M. A., & Jao, W.-C. 2004, AJ, 128, 437

Henry, T. J. 2004, ASP Conf. Ser. 318: Spectroscopically and Spatially Resolving the Com- ponents of the Close Binary Stars, 318, 159

Henry, T. J., Backman, D. E., Blackwell, J., Okimura, T., & Jue, S. 2003, The Future of Small Telescopes In The New Millennium. Volume III - Science in the Shadow of Giants, 111 – 23 –

Henry, T. J., Franz, O. G., Wasserman, L. H., Benedict, G. F., Shelus, P. J., Ianna, P. A., Kirkpatrick, J. D., & McCarthy, D. W., Jr. 1999, ApJ, 512, 864

Henry, T. J., Ianna, P. A., Kirkpatrick, J. D., & Jahreiss, H. 1997, AJ, 114, 388

Henry, T. J., Kirkpatrick, J. D., & Simons, D. A. 1994, AJ, 108, 1437

Henry, T. J., Subasavage, J. P., Brown, M. A., Beaulieu, T. D., Jao, W.-C., & Hambly, N. C. 2004, AJ, 128, 2460

Henry, T. J., Walkowicz, L. M., Barto, T. C., & Golimowski, D. A. 2002, AJ, 123, 2002

Honeycutt, R. K. 1992, PASP, 104, 435

Ianna, P. A., & Bessell, M. S. 1986, PASP, 98, 658

Jao, W.-C., Henry, T. J., Subasavage, J. P., Bean, J. L., Costa, E., Ianna, P. A., & M´endez, R. A. 2003, AJ, 125, 332

Jao, W.-C., Henry, T. J., Subasavage, J. P., Brown, M. A., Ianna, P. A., Bartlett, J. L., Costa, E., & M´endez, R. A. 2005, AJ, 129, 1954

Kalas, P., Liu, M. C., & Matthews, B. C. 2004, Science, 303, 1990

Landolt, A. U. 1992, AJ, 104, 372

L´epine, S., & Shara, M. M. 2005, AJ, 129, 1483

Luyten, W. J. 1979, LHS Catalogue (Minneapolis: Univ. of Minnesota Press)

Patterson, R. J., Ianna, P. A., & Begam, M. C. 1998, AJ, 115, 1648

Perryman, M. A. C., et al. 1997, A&A, 323, L49

Reid, I. N. 2003, AJ, 126, 2449

Reid, I. N., et al. 2003, AJ, 125, 354

Reid, I. N., et al. 2003, AJ, 126, 3007

Reid, I. N., et al. 2004, AJ, 128, 463

Reid, I. N., & Cruz, K. L. 2002, AJ, 123, 2806

Reid, I. N., & Gizis, J. E. 1997, AJ, 113, 2246 – 24 –

Reid, I. N., Kilkenny, D., & Cruz, K. L. 2002, AJ, 123, 2822

Reid, I. N., Kirkpatrick, J. D., Gizis, J. E., Dahn, C. C., Monet, D. G., Williams, R. J., Liebert, J., & Burgasser, A. J. 2000, AJ, 119, 369

Rodgers, A. W., & Eggen, O. J. 1974, PASP, 86, 742

Scholz, R.-D., Irwin, M., Schweitzer, A., & Ibata, R. 1999, A&A, 345, L55

Scholz, R.-D., Meusinger, H., & Jahreiß, H. 2001, A&A, 374, L12

Smart, R. L., et al. 2003, A&A, 404, 317

Subasavage, J. P., Henry, T. J., Hambly, N. C., Brown, M. A., & Jao, W.-C. 2005, AJ, 129, 413

Subasavage, J. P., Henry, T. J., Hambly, N. C., Brown, M. A., Jao, W.-C., & Finch, C. T. 2005, AJ, 130, 1658

Teegarden, B. J., et al. 2003, ApJ, 589, L51

Tinney, C. G. 1996, MNRAS, 281, 644

Tinney, C. G., Burgasser, A. J., & Kirkpatrick, J. D. 2003, AJ, 126, 975

Tinney, C. G., Reid, I. N., Gizis, J., & Mould, J. R. 1995, AJ, 110, 3014

van Altena, W. F., Lee, J. T., & Hoffleit, D. 1995, VizieR Online Data Catalog, 1174, 0

Vrba, F. J., et al. 2004, AJ, 127, 2948

Weis, E. W. 1984, ApJS, 55, 289

Weis, E. W. 1986, AJ, 91, 626

Weis, E. W. 1987, AJ, 93, 451

Weis, E. W. 1988, AJ, 96, 1710

Weis, E. W. 1991, AJ, 101, 1882

Weis, E. W. 1991, AJ, 102, 1795

Weis, E. W. 1993, AJ, 105, 1962

Weis, E. W. 1994, AJ, 107, 1135 – 25 –

Weis, E. W. 1996, AJ, 112, 2300

This preprint was prepared with the AAS LATEX macros v5.2. – 26 –

Fig. 1.— The nightly astrometric residuals in RA and DEC are shown for the binary star system, GJ 54 AB, in the top two panels, after fitting for parallax and proper motion. The 1.1 year orbital perturbation in the photocenter makes it difficult to fit the 1.0 year period of the parallactic motion, resulting in relatively high RMS residuals of 9.1 and 13.6 mas in RA and DEC, respectively. Some seasonal trends are seen in the data. The much lower residuals for the presumed single star, GJ 300, are shown in the bottom two panels, with

RMS values of only 3.6 mas in both RA and DEC. For GJ 300, a 10 MJup planet in a face-on circular orbit with a period of six years would cause the photocenter to trace a circle with a semimajor axis of 12 mas, clearly ruled out by the data. The semimajor axis expands to 19 mas for the same object in a 12 year orbit, which is twice the time coverage of the astrometric series currently available. For both systems, points with circles around them indicate nights on which only one frame was taken, and representative errors have been assigned.

Fig. 2.— The RECONS sample of objects in systems with trigonometric parallaxes larger

than 100 mas and errors less than 10 mas is plotted on an HR Diagram, using MV vs. V −K. Noteworthy stars are labeled, including the four known subdwarfs (triangles), the three ∼10 Myr old objects AU Mic and AT Mic A and B, and some of the recently discovered members of the 10 pc sample. Open points represent objects known in 2000 or discovered recently by other groups, and solid points represent RECONS discoveries. The three L dwarfs redder than V − K = 10 have all had distances measured since 2000.

Fig. 3.— A comparison of photometric distances and trigonometric distances is shown for objects within 20 pc with parallaxes published in the four CTIOPI papers reporting parallaxes to date. Open points are presumed single main sequence stars. Solid squares are known multiples with combined photometry. Three possible unnoticed doubles are shown with open squares — from left to right, GJ 555, ER 2, and LTT 6933, all from Jao et al. (2005), and LTT 6933 confirmed in Costa et al. (2005). The solid line represents equal photometric and trigonometric distances, and the dotted lines trace the region with distances differing by more than 20%. – 27 – – 28 – – 29 – – 30 –

Table 1. Ground-Based Trigonometric Parallaxes for Stars within 10 Parsecs Since YPC.a

Reference Observing # Total # New 10 pc New 10 pc Method Parallaxesb Systemsc Systemsc

Other Parallax Programs

Tinney (1996) optical CCD 13 1 LP 944-020 (= BRI 0337-3535) Dahn et al. (2002) optical CCD 28 2 2MA 0036+1821, 2MA 1507-1627 Reid et al. (2003a) optical CCD 1 1 2MA 1835+3259 Vrba et al. (2004) IR array 40 3 2MA 0415-0935, 2MA 0727+1710, 2MA 0937+2931

TOTAL (other programs) 82 7

RECONS Efforts

Henry et al. (1997) phot plates 1 1 GJ 1061 Jao et al. (2005) optical CCD 46 5 GJ 754, GJ 1068, GJ 1123, GJ 1128, DEN 1048-3956d Costa et al. (2005) optical CCD 31 2 GJ 2005 ABC, LP 647-013 Costa et al. (2006, AJ in press) optical CCD 31 1 DEN 0255-4700 this paper optical CCD 27 20 see Table 3

TOTAL (RECONS) 136 29

apapers listed describe new systems within 10 pc having parallax errors less than 10 mas breference star parallaxes not included in counts

c error in weighted mean of all πtrig must be less than 10 mas dalso reported in Costa et al. (2005) from CTIOPI at the 1.5 m – 31 – (16) # rel SpType 1.88 12 M4.5 V 1.84 12 M5.0 V 1.89 12 M3.5 V 3.11 12 M2.5 V 1.73 12 M1.5 V 1.26 12 M3.0 V 0.76 10 M8.5 VJ 1.57 12 M3.5 V 1.15 12 M5.5 V 1.86 12 M4.0 V 1.71 12 M5.5 V 1.73 12 M5.0 V 0.88 12 M6.0 V 0.69 12 M4.5 VJ 0.86 12 M6.0 VJ 0.98 12 M3.5 V 0.81 12 M4.5 V 0.93 12 M5.0 VJ 1.43 12 M3.0 V 1.09 12 M4.5 V 1.52 12 M4.5 V 0.61 12 M5.5 V 0.75 12 M7.0 V 1.04 12 M3.5 VJ 0.77 12 M3.0 VJ 1.49 12 M5.0 V 1.27 12 M3.0 VJ ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± phot 1 d 00 .44 9.70 6.15 3.56 12.25 s 2 9.33 69J 4.40 25 10.69 .33 11.38 .04 5.10 .13J 4.90 8.05 9.47 8.55 10.99 8.34 9.22 8J 7.92J 5.46 97J 8.51J 4.64 J 7.61J 7.28J 6.02 J 6.86J 6.59J 8.21 1J 6.56J 6.29J 6.43 JHK (mag) # nts # frm σ filter π ect, i.e. joint Not 10 Parsec Members New 10 Parsec Members Known 10 Parsec Members # nts KC I KC R Table 2. Photometric and Spectroscopic Results. J V (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Name1 Name2 Note. — J: photometry or spectral type is for more than one obj LHS 3746 11.76 10.56 9.04 4 V 0.013 33 213 7.60 7.02 6.72 8.14 SCR 1845-6357 AB 17.40J 15.00J 12.46J 5 I 0.004 18 117 9.54J 8. GJ 1207 LHS 3255 12.25 11.00 9.43 5 V 0.263v 27 124 7.97 7.44 7.1 WT 460 15.63 13.90 11.78 3 I 0.008 28 151 9.67 9.04 8.62 7.33 LHS 337 12.75 11.44 9.74 3 R 0.006 10 50 8.17 7.76 7.39 9.19 SCR 1138-7721 14.78 13.20 11.24 4 I 0.004 12 59 9.40 8.89 8.52 9 LHS 2206 G 042-024 14.02 12.63 10.85 3 R 0.011 20 118 9.21 8.60 8 LHS 288 13.90 12.31 10.27 3 R 0.007 13 68 8.49 8.05 7.73 6.90 LHS 2090 16.10 14.11 11.84 2 I 0.007 15 71 9.44 8.84 8.44 5.67 G 089-032 AB LTT 17993 13.25J 11.81J 9.97J 4 R 0.009 32 215 8.18 G 041-014 ABC 10.92J 9.67J 8.05J 3 V 0.009 22 160 6.51J 5.97J 5. SCR 0630-7643 AB 14.82J 13.08J 11.00J 4 I 0.005 12 69 8.89J 8.2 GJ 300 LHS 1989 12.15 10.85 9.22 3 V 0.016v 31 191 7.60 6.96 6.71 LP 771-095 BC LP 771-096 11.37J 10.13J 8.58J 4 V 0.043 21 98 7.1 G 099-049 LTT 17897 11.31 10.04 8.42 4 V 0.012 23 145 6.91 6.31 6 LHS 1723 12.22 10.87 9.18 5 V 0.020v 27 207 7.62 7.07 6.74 6.34 LP 771-095 A 11.22 10.07 8.66 4 V 0.012 22 109 7.29 6.77 6.50 8.9 LHS 1610 G 006-039 13.85 12.42 10.66 5 V 0.024v 23 128 8.93 8.38 GJ 1061 LHS 1565 13.09 11.45 9.46 5 R 0.010 27 186 7.52 7.02 6.61 SO 0253+1652 15.14 13.03 10.65 3 I 0.006 16 95 8.39 7.88 7.59 4. LHS 1302 G 159-003 14.49 13.00 11.17 5 R 0.021v 26 141 9.41 8.84 GJ 54 ABGJ 1001 A LHS 1208 9.82J LHS 102 8.70J 7.32J 12.86 5 11.64 10.10 V 2 0.015 R 20 0.010 149 6.00J 5.41J 14 5 61 8.60 8.04 7.74 APMPM J0237-5928 14.47 12.96 11.08 5 R 0.013v 27 166 9.28 8.70 GJ 633 LHS 3233 12.66 11.55 10.18 4 V 0.007 14 64 8.89 8.31 8.05 1 GJ 2130 A CD -32 13297 10.50 9.49 8.34 4 V 0.009 12 71 7.11 6.53 6. GJ 2130 BC CD -32 13298 11.51J 10.34J 8.86J 4 V 0.010 13 113 7.46 Table 3. Astrometric Results.

R.A. Decl. π(rel) π(corr) π(abs) µ P.A. Vtan −1 −1 Name J2000.0 J2000.0 Filter Nsea Nfrm Coverage Years Nref mas mas mas mas yr deg km sec notes (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

New 10 Parsec Members LHS 1302 01 51 04.09 −06 07 05.1 R 7c 141 1999.71−2005.96 6.25 6 100.14±1.89 0.64±0.06 100.78±1.89 597.1±0.9 115.5±0.17 28.1 APMPM J0237-5928 02 36 32.46 −59 28 05.7 R 7c 160 1999.64−2005.96 6.32 6 102.38±1.11 1.34±0.11 103.72±1.12 724.9±0.5 52.2±0.08 33.1 SO 0253+1652 02 53 00.89 +16 52 52.7 I 3c 95 2003.53−2005.88 2.35 6 258.48±2.68 2.15±0.21 260.63±2.69 5106.8±3.1 138.2±0.07 92.9 a LP 771-095 A 03 01 51.39 −16 35 36.0 V 6c 109 1999.64−2005.96 6.33 5 145.44±2.92 0.95±0.14 146.39±2.92 477.0±1.4 234.4±0.34 15.4 b LP 771-095 BC 03 01 51.04 −16 35 31.0 V 6c 99 1999.64−2005.96 6.15 5 138.75±4.99 0.95±0.14 139.70±4.99 479.4±2.5 235.5±0.59 16.3 LHS 1610 03 52 41.76 +17 01 04.3 V 7c 128 1999.71−2005.96 6.25 6 100.24±2.07 1.33±0.12 101.57±2.07 767.0±1.0 146.1±0.15 35.8 c LHS 1723 05 01 57.43 −06 56 46.5 V 7c 207 1999.81−2005.95 6.14 11 186.20±1.25 1.72±0.19 187.92±1.26 769.4±0.7 226.9±0.10 19.4 G 099-049 06 00 03.52 +02 42 23.6 V 7c 145 1999.91−2005.96 6.06 7 189.43±1.82 1.50±0.50 190.93±1.89 312.5±0.8 97.5±0.24 7.8 d SCR 0630-7643 AB 06 30 46.61 −76 43 09.0 I 3c 69 2003.94−2005.97 2.03 8 112.07±1.84 2.09±0.19 114.16±1.85 455.7±3.0 356.9±0.53 18.9 G 089-032 AB 07 36 25.13 +07 04 43.1 R 7c 216 1999.91−2005.95 6.05 13 115.67±0.97 0.93±0.08 116.60±0.97 396.9±0.5 143.0±0.14 16.1 GJ 300 08 12 40.88 −21 33 06.8 V 7c 191 1999.91−2005.95 6.05 8 122.40±0.90 3.20±0.37 125.60±0.97 698.8±0.5 178.7±0.06 26.4 e G 041-014 ABC 08 58 56.33 +08 28 26.0 V 7c 159 1999.97−2005.96 5.99 5 145.40±1.97 2.26±0.22 147.66±1.98 502.7±0.9 130.0±0.20 16.1 LHS 2090 09 00 23.55 +21 50 04.8 I 4c 71 2002.28−2005.20 2.92 9 155.98±2.67 0.89±0.06 156.87±2.67 773.9±2.2 221.2±0.33 23.4 LHS 2206 09 53 55.19 +20 56 46.8 R 7c 118 2000.06−2005.97 5.91 6 107.89±2.30 0.50±0.03 108.39±2.30 522.6±1.0 321.1±0.22 22.9 LHS 288 10 44 21.23 −61 12 35.6 R 5s 63 2000.06−2005.13 5.07 10 207.73±2.73 1.22±0.16 208.95±2.73 1642.9±1.2 347.7±0.07 37.3 f SCR 1138-7721 11 38 16.76 −77 21 48.5 I 3s 59 2003.23−2005.33 2.09 11 120.25±2.91 2.02±0.28 122.27±2.92 2147.6±5.1 286.9±0.25 83.3 LHS 337 12 38 49.10 −38 22 53.8 R 4s 50 2002.28−2005.47 3.19 14 155.37±1.99 1.41±0.14 156.78±1.99 1464.3±1.8 206.4±0.13 44.3 WT 460 14 11 59.94 −41 32 21.3 I 6c 152 2000.14−2005.57 5.43 10 105.58±1.51 1.83±0.13 107.41±1.52 690.4±0.9 260.0±0.12 31.0 GJ 1207 16 57 05.73 −04 20 56.3 V 7c 124 1999.62−2005.71 6.09 10 113.36±1.44 2.03±0.44 115.39±1.51 608.5±0.8 127.1±0.15 25.0 g SCR 1845-6357 AB 18 45 05.26 −63 57 47.8 I 3c 117 2003.24−2005.81 2.57 12 258.29±1.11 1.16±0.08 259.45±1.11 2664.4±1.7 76.6±0.06 48.7 h LHS 3746 22 02 29.39 −37 04 51.3 V 7c 213 1999.71−2005.88 6.17 7 133.14±1.20 1.15±0.11 134.29±1.31 820.7±0.8 105.1±0.09 29.0 Known 10 Parsec Members GJ 54 AB 01 10 22.90 −67 26 41.9 V 6c 149 2000.57−2005.96 5.39 8 139.28±3.32 1.92±0.62 141.20±3.38 682.5±2.1 33.1±0.34 22.9 i GJ 1061 03 35 59.71 −44 30 45.4 R 7c 186 1999.62−2005.95 6.33 7 270.98±1.34 0.94±0.08 271.92±1.34 826.3±0.8 117.7±0.10 14.4 j Not 10 Parsec Members GJ 1001 A 00 04 36.46 −40 44 02.7 R 5s 61 1999.64−2005.95 6.32 6 75.83±3.97 1.03±0.07 76.86±3.97 1627.0±1.8 156.7±0.12 100.3 k GJ 633 16 40 45.26 −45 59 59.3 V 6s 62 1999.64−2005.71 6.07 10 41.78±1.75 2.71±0.57 44.49±1.84 528.4±1.4 136.6±0.30 56.3 l GJ 2130 A 17 46 12.75 −32 06 09.3 V 5s 71 1999.64−2005.71 6.07 8 68.85±2.55 2.60±0.68 71.45±2.64 277.8±2.4 196.2±0.87 18.4 m GJ 2130 BC 17 46 14.42 −32 06 08.5 V 5s 113 1999.64−2005.71 6.07 8 67.63±1.93 2.60±0.68 70.23±2.05 277.4±1.8 196.4±0.66 18.7 – 33 – med to epoch 2000.0 using the proper motions and t’s coordinates were extracted from 2MASS and then transfor 7.53 mas in HIP ± 11.29 mas in HIP 10.1 mas in YPC and 122.86 38.04 mas in HIP 10.1 mas in YPC 11.4 mas in YPC 13.6 mas in YPC 15.0 mas in YPC 11.3 mas in YPC 5.2 mas in Henry et al. (1997) 13.7 mas in YPC ± ± 13.8 mas in YPC 23 mas in Deacon et al. (2005) 90 mas in Teegarden et al. (2003) ± ± ± ± ± ± ± ± ± ± ± parallax of 161.77 parallax of 92.97 parallax of 186.2 parallax of 282 parallax of 104.7 parallax of 104.4 parallax of 410 parallax of 70.0 parallax of 169.9 parallax of 222.5 parallax of 273.4 parallaxes of 120.5 parallax of 104.0 Note. — coordinates are epoch and equinox 2000.0; each targe b a c d g f e h i m j k l position angles listed here